This proposed CTDD program will help translate the enormous resource of high- throughput cancer genome characterizations into functionally validated, cancer-genotype based therapeutic targets. Over the past several years, members of this project have developed powerful tools and strategies for functionally annotating cancer genomes. These have enabled the identification and validation over fifty cancer genes, several of which are compelling therapeutic targets. Here, we propose to synthesize and optimize those strategies into a unified blueprint that can be applied across many tumor types. At the core of our philosophy is the use of powerful computational tools to narrow the extent of the genome that must be surveyed functionally. This enables the use of more precise human and mouse models to assess drivers and dependencies and approaches to combinatorial interactions that could not be carried out on a genome wide scale. We will apply these tools to select cancer types and gene sets in order to identify new oncogenic drivers, genotype-specific cancer dependencies, and to test strategies for the systematic identification of targets for combination therapies. We envision the impact of our project not only as the identification of genomically informed targets for several tumor types but also as providing tools and strategies that can be used throughout the consortium and the community.